Statistical and Scientific Considerations Concerning the Interpretation, Replicability, and Transportability of Research Findings

Abstract

To advance scientific understanding of disease processes and related intervention effects, study results should be free from bias and replicable. More broadly, investigators seek results that are transportable, meaning applicable to a perceived study population and also in other environments and populations. We review fundamental statistical issues which arise in the analysis of observational data from disease cohorts and other sources and discuss how these issues affect the transportability and replicability of research results. Much of the literature focuses on estimating average exposure or intervention effects at the population level, but we argue for more nuanced analyses of conditional effects that reflect the complexity of disease processes.

A practical guide to estimating treatment effects in patients with rheumatic diseases using real-world data

“…By adapting the codes to other variables and extended models, the practical guide could serve as a template for hands-on analysis of treatment effects using real-world data. It may facilitate aims outlined in guidelines and recommendations for analyzing and reporting observational studies in rheumatology [ 3 – 5 , 9 ].…”

“…A recent review on statistical concerns in rheumatology research thoroughly outlines issues that arise in the analysis of disease cohorts and other sources. It emphasizes the need for careful analysis of observational data [ 9 ]. If the study design does not address causality, for example, through randomization, causal inference methods can be useful and may provide additional insights compared to classical statistical methods such as basic tests and regression analysis [ 10 ].…”